AnovaX: A Local, Multi-Agent Voice Assistant with LLM Planning and Adaptive Recovery

AnovaX is a new voice assistant that runs entirely on your computer, not in the cloud. Unlike Siri or Alexa, it doesn't ship raw audio off the machine. Instead, it treats the desktop itself as its action surface. A single Python process wires together a wake-word gate, a speech pipeline, an LLM planner (Gemini) that emits a JSON plan of tool calls, a whitelist-and-denylist safety layer, and a multi-agent orchestrator that translates each plan into typed child agents on a bounded thread pool. The system also features adaptive recovery, allowing it to handle errors and retry tasks gracefully.

This is a significant step for privacy and efficiency. Since AnovaX doesn't rely on cloud processing, it can respond faster and keep your data secure. It also adapts to your workflow, making it easier to manage tasks without switching between multiple apps. Imagine asking it to draft an email, search for documents, and schedule a meeting—all without leaving your desktop environment.

If you're curious about trying a local voice assistant, check out open-source projects like Mycroft or Rhasspy. These tools offer similar local processing capabilities and can be a good starting point for experimenting with voice-controlled tasks on your own computer.